Locked by Moderator, Not on subject anymore,AIR Roadmap Update

[chris.campbell]

Hello AIR developers,

With the news today regarding Flash Player, I'm sure many of you have questions regarding AIR and its future roadmap.  Let me start by saying that today's announcement was not about AIR, and instead focuses entirely on Flash Player and the browser plugin environment.  Adobe remains committed to AIR and we believe it continues to be a great desktop and mobile development platform.

Many of you have asked for a roadmap update.  We hope to have our official Flash Runtime roadmap updated soon, but until then I wanted to share some of the features we'd like to accomplish in our upcoming releases.  As always, this list may change as we receive feedback from the community.

• Support for 64-bit AIR – Windows Captive Only
• AIR SDK installer on Windows
• Desktop async texture upload
• DirectX11 support for AIR desktop
• Increase the GPU memory texture limit
• Improved monitor and resolution settings for AIR Desktop
• GPU render mode for AIR Desktop
• Support for the latest SDK on iOS 11 and Android O
• ASTC Support for mobile
• VR/AR support for AIR apps
• ANE support for Swift on iOS
• METAL bindings for Stage3D
• Motion detection for Android devices

We've also been following a feature request thread on the Starling forums.  We wanted to get your feedback on some of the items outlined by the community.  If you'd like to provide additional input, please take a minute and take this three question survey so we can better understand what folks would like to see in future releases.

Community AIR Feature Requests​ Survey

[SCORMpool]

Without Flash air will be one of many cross platform development tools. Check http://haxe.org and based on it http://www.openfl.org/​ for example. Haxe already has html/js output format and many more on top of it. Now it will be really hard for Air take leading place. Sad Adobe dint look at option make their own browser optimized for flash. Good luck!

[Greg_G]

Will Adobe be releasing a official statement about the longevity of AIR past 2020? I do not want to sound mean, but we really need more than a forum post to chart our course forward with our products that drive our revenue. 2.5 years is not much time in enterprise solution development time.

[Born2code]

I agree, we do need at least a 5 year commitment from Adobe on AIR.

[W.P. Stuurman]

Actually, either a 5 year commitment OR the assurance AIR will be open sourced (to e.g. Apache) if Adobe should at some stage not continue it.

[Born2code]

Absolutely, I 100% agree, AIR should be open sourced if Adobe drops it in the future.

We have millions of dollars invested behind it !!!

[Joseph_Saade]

That is some great news. But from a developer point of view we would like to see some other stuff:

- Better support and bug tracking system. Maybe forum anf bugbase could be under one roof to facilitate issues.

- Fix old Air bugs ( issues with camera, keyboard, text , text input and webviews on mobile devices). Probably also make stagevideo work with rtmp across mobile devices. Other old issues are video in gpu mode does not always work.

- Ability to mix and match renderers dynamically and not just cpu/gpu/direct mode. example stage3d is always behind native display list no matter what so some edge cases like native video + starling on top does not work.

Thanks

[Guest]

Good news! Thanks!

Adobe AIR is by far the best available tool for cross platform development.

I hope, that Adobe will continue to move forward with this technology.

I think however, it could be a bit more present in the communication of Adobe, so that people know, that it is there and available.

It's too good to be missed...

[el111]

For those talking about Vector rendering for Stage3d / Starling - this may be of interest

Termite – a vector renderer for stage3d (and possibly other targets) « flashing in public

The author is looking for feedback and to measure interest. I personally think it looks extremely impressive.

On the original post from Chris Campbell... I think there are some great looking features in the pipeline.

Out of interest what are people ranking as #1 on the additional survey offered. I thought "OpenGL Enhancements like Access Stage3D textures directly from an ANE, Update to Open GL ES 6+, Ability to use OpenGL instead of DirectX" sounded intriguing.

Regarding asking for 10 year guarantees. Is anyone able to say any alternative tech will 100% be around and relevant in 10years?

Take your pick. Drop them a line on their support forum asking them to put in writing a binding guarantee they will be around in 10 years.

This is the tech industry we are talking about.

What I read in the original post is that the EOL for Flash player is 2020 and the announcement had no bearing on AIR. And I was happy with that.

[ASWC]

Of course the announcement has direct bearing on AIR, the Flash player is the backbone of AIR, the "engine" if you will. AIR will continue to develop its own features within its own scope but it will do it based around a core engine (FP) that will never evolve anymore. After 2020 the AIR platform by definition will have a expiration date, period and no amount of wishful thinking will change that.

[Colin Holgate]

It's Flash Player, the browser plugin, that will end of life. SWF will continue to be advanced for the purposes of animation to video, and for use in AIR apps.

[ASWC]

Animation and video? For what purpose? (if not to run in a browser in what platform is this going to run?)

As for AIR yeah maybe but developing flash technology just for AIR doesn't sound like something Adobe is gonna want to do. We'll see after 2020 but Adobe as never once surprised me ....

[Colin Holgate]

Animation would mainly be for television or for YouTube and Vimeo videos.

I've seen figures of 10% of apps submitted to app stores being done as AIR apps, so it's still a significant market for Adobe, even discounting all the possible desktop AIR apps.

[ASWC]

10% seems high but I take your words for it. Where did you find those figures?

[Joseph Labrecque]

The most direct bearing on AIR this announcement has is more resources for AIR and freeing the core technology from restrictions present in the web browser.

[ASWC]

Would be ideal sure, is this really what's gonna happen? Let me answer with a question: Did Adobe ever took a decision or made a move that would ensure the future of Flash technology?

[Colin Holgate]

Yes, they withdrew support for Director.

[JoãoCésar]

Animate CC has the best integration of art, animation, and code of all platforms I ever used and I heard of.

There should be a way of creating AIR apps powered by Starling, Feathers, and any other framework graphically inside of the IDE.

There are tons of pure code solutions but there isn't a platform that allows the user to start a project really from the scratch, from the art to the final developed product, in such an easy workflow like Animate.

[SCORMpool]

There is no warranty that  any technology will be available as long as you want it. On my experience do it by yourself paradigm make more and more sense. Browsers nothing but sand box where big corporations let you play, you always be restricted and bumped  on cross browser/versions compatibility. Only C/C++ level gives you true freedom. If you have option use app or html version of app, what would you choose? I don't think it will be some kind magical moment in time when html app suddenly became better then native apps.

Did few experiments, trying to understand if it even possible archive the same level using open sourced libraries:

https://play.google.com/store/apps/details?id=com.codeservice.app.luastudio

https://appsto.re/ca/x92nfb.i

[tkcbk57594040]

I like Flash, I like ActionScript, and I like AIR.

I hope Adobe continues to support him because he's a great technology.

It is web HTML, Flash Player is a great plugin, but I hope it has always been there, until one day HTML in all aspects beyond Flash Player (but I think this is very rare, the development of HTML itself is contrary to the original definition for HTML).

TKCB

[Joseph_Saade]

I have a cloud license and develop on animate or FB4.7.

I have worked on a lot of mobile and desktop projects and I can say one thing, AIR, in general, has an identity crisis.

I am not sure if it is targetted towards apps or games. Both seem half baked, missing the last touches. Everytime I have an app, I feel I have to recreate the UI and test everything on every device to make sure simple stuff like text renders fine.

And when I have a game, I usually use starling but lose the flexibility if Animate.

I can list a lot of half baked features example remote notification support on iOS only (?)

In all it is an old and mature platform, so some minor issues that should work out of the box and do not are always weird.

From my point of view, having AIR as open source would solve all these compatibility and weird issues and enable me as a developer to fix minor bugs without relying too much on the weird bugbase and blackbox.

[TeotiGraphix]

Cross compilers are nothing new and well, I had that dream back in 2011/2012 I wrote the cross compiler that Apache Flex is using right now in FlexJS, FalconJX.

See the thing is, with me, the AS3 language is not evolving. I spent over 1000 hours on that code base and unit tests in Java. Why? Because Adobe screwed devs(Novemebr 2011), go read my bio on Apache Flex's site, "I am writing the compiler to put the last nail in the Flash Player". Something along those lines.

So what I am saying is, if I had a wish it would be to get new language features.

Ah yes, but there is a catch, see Adobe wasn't stupid when they trashed Flex, they forked the compilers when they donated Falcon. So now, you have two projects that support AS3 but can never have parity with each other.

So on one hand you have AIR and it's compiler, then you have Apache with it's Falcon/FalconJX compiler.

Now with the nail, the one nasty factor is byte code. Since AIR is still using Flash runtime, any real push from Apache Flex is stomped on because we cannot add features or adjust byte code for AIR.

I still use AIR and have a popular drum machine on Android, I love it but the fact I could not donate all my compiler experience to push AS3 into the future sucks bad. (why I am not active in the project right now)

[ASWC]

AS3 is pretty good but AS NEXT was going to be awesome ... Adobe pulled the plug on that too.

[TeotiGraphix]

Agreed, I was on that boat the whole time.

The fact is, AS3 is just grammar with an AST tree. That AST tree represents the whole language. It can be manipulated any way and extended.

The Falcon compiler had such a great multi-threaded architecture that could easily be extended to add new language features.

As I said, Adobe put the last nail in AS3, Flash Player and AIR when they gave Apache half of the project. What good is a compiler when AIR uses the ASC2 compiler that is closed source?

It's a joke, and once I took my rose colored glasses off and realized Adobe was doing it as a marketing thing for there "love of the community"(the donation), I had to realize that no matter what talent the OS community had to extend a compiler for the ActionScript language, as long as Adobe has the keys to the AIR compiler, it's futile to dream of any future.

[rewb0rn]

Would it be possible in theory to release the ASC 2.0 compiler to open source, or are licensing agreements preventing Adobe from considering that? Because if it was open source, people could start building an HTML / WebGL compilation target, no? It is still our major concern in the company that with Air there is no future in the browser and that will force us away to some other technology.

[TeotiGraphix]

No, ASC2 is connected to Animate. Looking back at it now, 2011 Adobe knew Flash was dead and the first thing they needed to do was get rid of the Flex Developers. So they donated "a" compiler and a half baked cross compiler FalconJS to Apache.

FalconJS was such a mess I decided to write FalconJX which was a simple recursive AST traverse and emitter design. It actually had a default AS3 emitter that I used in unit tests.

The fact is, FalconJX can emit anything you want. Where the trouble is, is the byte code. As I said previously, if the Flash run-time in C++ remains closed, the ASC compiler will be closed, they are bound together like peanut butter and jelly.

Back eons ago in 2011 we had a project called Randori(dumb name) but it was a POC of AS3 to JS/WebGL, Renaun actually got Hungry Heros to run using the compiler I worked on. That version of the compiler I actually then merged back into FalconJX at Apache.

Anyway, TLDR, Adobe must open source the run-time, you need the compiler AND rum-time to be OS for AIR to be completely community driven.